| Index: source/data/brkitr/word.txt
|
| diff --git a/source/data/brkitr/word.txt b/source/data/brkitr/word.txt
|
| index c74da4cdd9c5867e1b8152d3b76d9e4f39ec90b5..f89a2fe74526fc0e521f942b5f50957377a1a040 100644
|
| --- a/source/data/brkitr/word.txt
|
| +++ b/source/data/brkitr/word.txt
|
| @@ -35,16 +35,10 @@ $Hebrew_Letter = [\p{Word_Break = Hebrew_Letter}];
|
| $ALetter = [\p{Word_Break = ALetter}];
|
| $Single_Quote = [\p{Word_Break = Single_Quote}];
|
| $Double_Quote = [\p{Word_Break = Double_Quote}];
|
| -# Remove two full stop characters from $MidNumLet and add them to $MidNum
|
| -# to break a hostname into its components at the cost of breaking
|
| -# 'e.g.' and 'i.e.' as well.
|
| -# $MidNumLet is used in rules 6/7 (rules of our interest) and rules 11/12.
|
| -# Because it's OR'd with $MidNum in rules 11/12, rules 11/12 are not affected
|
| -# while rules 6/7 are reverted to the old behavior we want.
|
| -$MidNumLet = [[\p{Word_Break = MidNumLet}] - [\u002E \uFF0E]];
|
| +$MidNumLet = [\p{Word_Break = MidNumLet}];
|
| $MidLetter = [\p{Word_Break = MidLetter}];
|
| -$MidNum = [\p{Word_Break = MidNum}[\u002E \uFF0E]];
|
| -$Numeric = [\p{Word_Break = Numeric}[\uff10-\uff19]]; #includes fullwidth digits
|
| +$MidNum = [\p{Word_Break = MidNum}];
|
| +$Numeric = [\p{Word_Break = Numeric}];
|
| $ExtendNumLet = [\p{Word_Break = ExtendNumLet}];
|
|
|
| $Han = [:Han:];
|
|
|